| 325eefa2 |
/*
* DSP utils
* Copyright (c) 2000, 2001 Fabrice Bellard
* Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
*
* gmc & q-pel & 32/64 bit based MC by Michael Niedermayer <michaelni@gmx.at>
* |
| d18e243f |
* This file is part of FFmpeg. |
| 325eefa2 |
* |
| d18e243f |
* FFmpeg is free software; you can redistribute it and/or |
| 325eefa2 |
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
* |
| d18e243f |
* FFmpeg is distributed in the hope that it will be useful, |
| 325eefa2 |
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public |
| d18e243f |
* License along with FFmpeg; if not, write to the Free Software |
| 325eefa2 |
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/**
* @file
* DSP utils
*/
|
| c45f6295 |
#include "bit_depth_template.c" |
| 325eefa2 |
|
| 8a523cfa |
#if BIT_DEPTH == 8 |
| 325eefa2 |
/* draw the edges of width 'w' of an image of size width, height */
//FIXME check that this is ok for mpeg4 interlaced |
| c137fdd7 |
static void FUNCC(draw_edges)(uint8_t *p_buf, int p_wrap, int width, int height, int w, int h, int sides) |
| 325eefa2 |
{ |
| d18e243f |
pixel *buf = (pixel*)p_buf;
int wrap = p_wrap / sizeof(pixel); |
| 87ce8b49 |
pixel *ptr, *last_line; |
| 325eefa2 |
int i;
/* left and right */
ptr = buf;
for(i=0;i<height;i++) {
memset(ptr - w, ptr[0], w);
memset(ptr + width, ptr[width-1], w);
ptr += wrap;
}
/* top and bottom + corners */
buf -= w;
last_line = buf + (height - 1) * wrap;
if (sides & EDGE_TOP) |
| c90b9442 |
for(i = 0; i < h; i++) |
| 87ce8b49 |
memcpy(buf - (i + 1) * wrap, buf, (width + w + w) * sizeof(pixel)); // top |
| 325eefa2 |
if (sides & EDGE_BOTTOM) |
| c90b9442 |
for (i = 0; i < h; i++) |
| 87ce8b49 |
memcpy(last_line + (i + 1) * wrap, last_line, (width + w + w) * sizeof(pixel)); // bottom |
| 325eefa2 |
} |
| 8a523cfa |
#endif |
| 325eefa2 |
|
| 4a88d81c |
static void FUNCC(get_pixels)(int16_t *av_restrict block,
const uint8_t *_pixels,
int line_size)
{
const pixel *pixels = (const pixel *) _pixels;
int i;
/* read the pixels */
for(i=0;i<8;i++) {
block[0] = pixels[0];
block[1] = pixels[1];
block[2] = pixels[2];
block[3] = pixels[3];
block[4] = pixels[4];
block[5] = pixels[5];
block[6] = pixels[6];
block[7] = pixels[7];
pixels += line_size / sizeof(pixel);
block += 8;
} |
| 325eefa2 |
}
|
| 4a88d81c |
#if BIT_DEPTH == 8
static void FUNCC(clear_block)(int16_t *block)
{
memset(block, 0, sizeof(int16_t)*64);
}
static void FUNCC(clear_blocks)(int16_t *blocks)
{
memset(blocks, 0, sizeof(int16_t)*6*64);
} |
| 5cc26009 |
#endif |
| 325eefa2 |
|
| 9d40f710 |
#if BIT_DEPTH == 8 |
| e9d81735 |
#include "hpel_template.c" |
| 9d40f710 |
#endif |
| 2612c4dc |
#define PIXOP2(OPNAME, OP) \
static inline void FUNC(OPNAME ## _no_rnd_pixels8_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
int i;\
for(i=0; i<h; i++){\
pixel4 a,b;\
a= AV_RN4P(&src1[i*src_stride1 ]);\
b= AV_RN4P(&src2[i*src_stride2 ]);\
OP(*((pixel4*)&dst[i*dst_stride ]), no_rnd_avg_pixel4(a, b));\
a= AV_RN4P(&src1[i*src_stride1+4*sizeof(pixel)]);\
b= AV_RN4P(&src2[i*src_stride2+4*sizeof(pixel)]);\
OP(*((pixel4*)&dst[i*dst_stride+4*sizeof(pixel)]), no_rnd_avg_pixel4(a, b));\
}\
}\
\ |
| 87ce8b49 |
static inline void FUNC(OPNAME ## _no_rnd_pixels16_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \ |
| 325eefa2 |
int src_stride1, int src_stride2, int h){\ |
| 87ce8b49 |
FUNC(OPNAME ## _no_rnd_pixels8_l2)(dst , src1 , src2 , dst_stride, src_stride1, src_stride2, h);\
FUNC(OPNAME ## _no_rnd_pixels8_l2)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, h);\ |
| 325eefa2 |
}\
\ |
| 87ce8b49 |
static inline void FUNC(OPNAME ## _pixels8_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\ |
| 325eefa2 |
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
| 87ce8b49 |
/* FIXME HIGH BIT DEPTH */\ |
| 325eefa2 |
int i;\
for(i=0; i<h; i++){\
uint32_t a, b, c, d, l0, l1, h0, h1;\
a= AV_RN32(&src1[i*src_stride1]);\
b= AV_RN32(&src2[i*src_stride2]);\
c= AV_RN32(&src3[i*src_stride3]);\
d= AV_RN32(&src4[i*src_stride4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
a= AV_RN32(&src1[i*src_stride1+4]);\
b= AV_RN32(&src2[i*src_stride2+4]);\
c= AV_RN32(&src3[i*src_stride3+4]);\
d= AV_RN32(&src4[i*src_stride4+4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride+4]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
}\
}\
\ |
| 87ce8b49 |
static inline void FUNC(OPNAME ## _no_rnd_pixels8_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\ |
| 325eefa2 |
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
| 87ce8b49 |
/* FIXME HIGH BIT DEPTH*/\ |
| 325eefa2 |
int i;\
for(i=0; i<h; i++){\
uint32_t a, b, c, d, l0, l1, h0, h1;\
a= AV_RN32(&src1[i*src_stride1]);\
b= AV_RN32(&src2[i*src_stride2]);\
c= AV_RN32(&src3[i*src_stride3]);\
d= AV_RN32(&src4[i*src_stride4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x01010101UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
a= AV_RN32(&src1[i*src_stride1+4]);\
b= AV_RN32(&src2[i*src_stride2+4]);\
c= AV_RN32(&src3[i*src_stride3+4]);\
d= AV_RN32(&src4[i*src_stride4+4]);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x01010101UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
l1= (c&0x03030303UL)\
+ (d&0x03030303UL);\
h1= ((c&0xFCFCFCFCUL)>>2)\
+ ((d&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)&dst[i*dst_stride+4]), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
}\
}\ |
| 87ce8b49 |
static inline void FUNC(OPNAME ## _pixels16_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\ |
| 325eefa2 |
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
| 87ce8b49 |
FUNC(OPNAME ## _pixels8_l4)(dst , src1 , src2 , src3 , src4 , dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
FUNC(OPNAME ## _pixels8_l4)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), src3+8*sizeof(pixel), src4+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\ |
| 325eefa2 |
}\ |
| 87ce8b49 |
static inline void FUNC(OPNAME ## _no_rnd_pixels16_l4)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, const uint8_t *src4,\ |
| 325eefa2 |
int dst_stride, int src_stride1, int src_stride2,int src_stride3,int src_stride4, int h){\ |
| 87ce8b49 |
FUNC(OPNAME ## _no_rnd_pixels8_l4)(dst , src1 , src2 , src3 , src4 , dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\
FUNC(OPNAME ## _no_rnd_pixels8_l4)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), src3+8*sizeof(pixel), src4+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, src_stride3, src_stride4, h);\ |
| 325eefa2 |
}\
\ |
| 0ddca7d4 |
static inline void FUNCC(OPNAME ## _pixels8_xy2)(uint8_t *block, const uint8_t *pixels, ptrdiff_t line_size, int h)\ |
| 325eefa2 |
{\ |
| 87ce8b49 |
/* FIXME HIGH BIT DEPTH */\ |
| 325eefa2 |
int j;\
for(j=0; j<2; j++){\
int i;\
const uint32_t a= AV_RN32(pixels );\
const uint32_t b= AV_RN32(pixels+1);\
uint32_t l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
uint32_t h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
uint32_t l1,h1;\
\
pixels+=line_size;\
for(i=0; i<h; i+=2){\
uint32_t a= AV_RN32(pixels );\
uint32_t b= AV_RN32(pixels+1);\
l1= (a&0x03030303UL)\
+ (b&0x03030303UL);\
h1= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
pixels+=line_size;\
block +=line_size;\
a= AV_RN32(pixels );\
b= AV_RN32(pixels+1);\
l0= (a&0x03030303UL)\
+ (b&0x03030303UL)\
+ 0x02020202UL;\
h0= ((a&0xFCFCFCFCUL)>>2)\
+ ((b&0xFCFCFCFCUL)>>2);\
OP(*((uint32_t*)block), h0+h1+(((l0+l1)>>2)&0x0F0F0F0FUL));\
pixels+=line_size;\
block +=line_size;\
}\
pixels+=4-line_size*(h+1);\
block +=4-line_size*h;\
}\
}\
\ |
| 87ce8b49 |
CALL_2X_PIXELS(FUNCC(OPNAME ## _pixels16_xy2), FUNCC(OPNAME ## _pixels8_xy2), 8*sizeof(pixel))\ |
| 325eefa2 |
|
| 87ce8b49 |
#define op_avg(a, b) a = rnd_avg_pixel4(a, b) |
| 325eefa2 |
#define op_put(a, b) a = b |
| 2612c4dc |
#if BIT_DEPTH == 8 |
| e9d81735 |
#define put_no_rnd_pixels8_8_c put_pixels8_8_c |
| 325eefa2 |
PIXOP2(avg, op_avg)
PIXOP2(put, op_put) |
| 2612c4dc |
#endif |
| 325eefa2 |
#undef op_avg
#undef op_put |